Databases Reference
In-Depth Information
and modifications to the ETL processing. This happened repeatedly, which
frustrated the project team members and the business group waiting for their
reports.
After weeks of running in circles, the project team took a moment to review
what was happening. It was determined that too many resources and time had
been wasted already and a long-term solution was not in sight. The project
team enlisted the help of the business champion to refine the project plan
by going back to develop a robust data model. This required invoking the
change control process. Key business analysts were assigned to help with the
modeling process. The model was successfully completed and well received
by the rest of the business community. Everyone is anxious for the database to
be built so that they can begin to use it.
Summary
Developing a sound data model up front sets the stage for success. Dimensional
modeling helps to organize data for reporting and analysis, often directly by
the business audience. In order to facilitate business use, it is important
that the model reflect the business perspective. The business dimensional
model diagrams help to capture this perspective. A proven process to develop
the dimensional model was also presented. The dimensional model can be
implemented in a variety of different technologies.
Useful information is uncovered during the modeling process and can
be used to bridge to subsequent stages of a data warehouse project. The
preliminary source-to-target data map captures details that will be needed to
develop the ETL system. Similarly, the Business Measures Worksheet content
is needed to set up the data delivery. Before discussing these subsequent
steps, the next chapter spends a bit more time looking at data-related topics,
including information management, data governance, and data quality.
